Make your own free website on Tripod.com

9 Testing

This section evaluates the simulation against the requirements in Section 5 and the software design specification in Section 6. Screenshots of the simulated impairments are shown in Appendix B.

9.1 Functional Requirements

Table 9.1 evaluates the simulation against the functional requirements in Section 5.2 and the software design specification in Sections 6.1 to 6.5:

Requirement Requirement Does the simulation satisfy the requirement?
Req 1: The software shall simulate the symptoms of one or more visual impairments.
Req 1.1 In Part   The software simulated myopia by blurring the displayed image. However, all objects were blurred to the same degree, irrespective of the distance from the viewer (see Req 2.2).
Req 1.2 In Part  The software simulated hyperopia by blurring the displayed image. However, all objects were blurred to the same degree, irrespective of the distance from the viewer (see Req 2.2).
Req 1.3 Yes The software simulated glaucoma by creating a tunnel vision effect. At a low severity, a large circular area of the display was unobscured. At a high severity a small circular area of the display was unobscured. The region outside the tunnel was black.
Req 1.4 Yes The software simulated retinitis pigmentosa by creating a tunnel vision effect. At a low severity, a large circular area of the display was unobscured. At a high severity a small circular area of the display was unobscured. The region outside the tunnel was black.
Req 1.5 Yes The software simulated macular degeneration by drawing a black patch to obscure an area of display at the centre of the field of vision. The size of the black patch increased with the severity.
Req 1.6 Yes The software simulated diabetic retinopathy by obscuring the display with black patches which increase in size with the severity. However, these black patches were all circles rather than irregular 'blobs'.
Req 1.7 Yes The software simulated two variations of hemianopia where the right side of the visual field was obscured and the macula was either 'spared' or 'involved' in the loss of vision.
Req 1.8 Yes The software simulated three variations of cataracts. The symptoms of the cloudy vision variation were simulated by creating a grey cloudy area at the centre of vision which increased in size as the severity increased. The symptoms of the yellow / brown variation were simulated by darkening the image and adding a yellow tint and the effect was increased with the severity. The double vision effect was applied using the accumulation buffer and increased as the severity increased.
Req 1.9 In Part The software provided a simple simulation of colour blindness by creating a greyscale image from the displayed output. However, the software did not simulate the less severe variations of colour blindness where red, orange and yellow colours are shifted in hue towards green, and purple and violet appear blue.
Req 2: The software simulation shall be dynamic.
Req 2.1 Yes The software simulated the movement of the field of vision by implementing a 'track to cursor' mode which allowed the user to select the centre of the field of vision for the simulated impairments by moving the mouse cursor.
Req 2.2 No The software did not simulate varying the focal length of the vision. The simulated impairments affected all objects in the same way, irrespective of the distance from the viewer.
Req 3: The user shall be able to interact with the simulation.
Req 3.1 Yes The system allowed the user to change the type of impairment via a popup menu or a keyboard shortcut.
Req 3.2 Yes The system allowed the user to adjust the severity of the impairment via a popup menu option or a keyboard shortcut.
Req 3.3 Yes The system allowed the user to select the centre of the field of vision via a popup menu option or a keyboard shortcut.
Req 3.4 No There was no option to change the focal distance of the simulated impairments (see Req 2.2).

Req 3.5

Yes The system allowed the user to save and recall default settings for the type and severity of the impairment and the position of the centre of the field of vision via a popup menu option.
Req 3.6 Yes  The system allowed the user to exit the simulation via an option on the popup menu or by pressing the 'Esc' key.
Req 4: The simulation shall run in true full screen mode.
Req 4.1 Yes The simulation operated in true full screen mode and used a window which had no title bar or borders and had the dimensions of the screen.
Req 5: The software shall simulate the use of a computer with a visual impairment.
Req 5.1 Yes In 'transparent' mode the simulated impairments were applied to the contents of the screen behind the main simulation window. The system also allowed the use of a simple drawing program and a web browser with the simulated impairments applied to the output.
Req 5.2 In Part  The system allowed the user to interact with the drawing program and web browser by using the mouse and keyboard, but did not allow interaction with other software applications.

Table 9.1 - Evaluating the Functional Requirements

The system simulated all of the visual impairments specified in Req 1, however, the simulations of myopia, hyperopia and colour blindness did not fully meet the more detailed requirements in the software design specification. All of the simulated impairments satisfed the first part of Req 2 by allowing the user to move the centre of the field of vision, but did not satisfy the second part of the requirement as there was no method provided to change the focal length of the vision. Most computer applications use two-dimensional graphics. Varying the focal length of the vision would have only a small effect on the use of a computer with a visual impairment.

The system allowed the user to interact with the simulation. The system did not simulate the effect of varying the focal distance so there was no menu option to change it. The system satisfied the remaining parts of Req 3. The system operated in true full screen mode, which satisfied Req 4. The system satisfied the first part of Req 5 but did not completely satisfy the second part of the requirement. The system simulated the use of a computer with a visual impairment by providing an interactive drawing program and web browser and applying the impairments to the displayed output.

9.2 Usability Requirements

Table 9.2 evaluates the simulation against the usability requirements in Section 5.3:

Requirement 

Does the simulation satisfy the requirement?
Req 6.1 Yes

 The system was consistent. The actions required to change the severity and the centre of the field of vision were the same for each of the impairments.

Req 6.2 Yes The system provided feedback by changing the mouse cursor while the display was updated.
Req 6.3  Yes The system did not provide undo and redo options, but did allow easy reversal of actions by allowing the user to reselect the type and severity of the simulated impairment via the popup menu and keyboard shortcuts.
Req 6.4 Yes The system provided keyboard shortcuts to allow the user to change the type and severity of the simulated impairment.
Req 6.5 Yes The system provided customisation of actions by allowing the user to save and recall default settings for the simulation (see Req 3.5).
Req 6.6 Yes The user interface consisted of a single popup menu which allowed the user to change the type and severity of impairment and the centre of the field of vision. Changing a system setting required the user to make a single selection from this menu. Documentation for the system included the layout of the popup menu (Appendix B) and the keyboard shortcuts (Table 8.7 and Table 8.8).

Table 9.2 - Evaluating the User Requirements

9.3 Performance Requirements

Table 9.3 evaluates the simulation against the performance requirements in Section 5.3:

Requirement 

Does the simulation satisfy the requirement?
Req 7.1 Yes  The simulation took on average no longer than 2 seconds to start. This was less than the 30 seconds specified in requirement.
Req 7.2  In Part  The simulations of glaucoma, macular degeneration, diabetic retinopathy and hemianopia all took on average less than 1 second to update the display. This was less than the 10 seconds specified in the requirement. The simulations of cataracts took between 1 and 8 seconds to update the display depending on the selected variation and severity which was less than the required 10 seconds. The simulations of myopia and hyperopia when applied at a low severity took approximately 15 seconds to update the display, and at a high severity more that 30 seconds to update the display. These were both longer than the required 10 seconds.

Table 9.3 - Evaluating the Performance Requirements

The simulations of myopia and hyperopia did not meet the performance requirements for the system. The relatively long times required to update the display for these simulations was due to the computation required to blur the image using the accumulation buffer. The blur operation applied the convolution kernel to every pixel in the image (Section 7.4.2). The time taken to update the display was therefore dependant on the resolution of the screen. The update times were measured by running the simulation on a screen with a resolution of 1024 x 768 pixels. Increasing the resolution of the screen would have resulted in even greater update times.

9.4 Summary

The simulations of glaucoma, retinitis pigmentosa, macular degeneration, diabetic retinopathy, hemianopia and cataracts all satisfied the functional requirements for the system. The simulation of myopia and hyperopia did not completely satisfy the functional requirements or the performance requirements of the system. The simulation of colour blindness also did not completely satisfy the functional requirements. All of the usability requirements for the system were satisfied. The parts of the simulation that did not meet the requirements highlight areas for further development. These are described in Section 11.3.


Stephen Ratcliffe. 2005
Department of Computer Science
University of York
Heslington
York
YO10 5DD